Session 3B Doyle, Rutherford, McDonnell

8/11/2019 Session 3B Doyle, Rutherford, McDonnell

1/22

`

Volunteering Participation inScotland: An Analysis of Linked

Social Survey and AdministrativeDataVoluntary Sector and VolunteeringResearch Conference 2014

Kathleen Doyle, Volunteer Scotland

Alasdair Rutherford, University of Stirling

Diarmuid McDonnell, University of Stirling


2/22

`

The squeeze on government spending has focussed policy makers attention on therole of voluntary participation in the provision of public services, leading to greaterfunding scrutiny and a need to understand how communities can participate insustainable public services. This policy shift towards co-production assumes thatlevels of volunteering and its distribution in all localities can meet this demand. Thispaper explores data on volunteering participation in Scotland to identify thedeterminants of volunteering supply. It examines both participation decisions andnumber of hours volunteered. The findings show that urban/rural differences persist

even after controlling for local area characteristics. While participation rates arehigher in rural areas, the intensity of individual volunteering is no different. 1

Keywords: Volunteering, Participation, Co-production, Urban/Rural

1 The project was funded by the Applied Quantitative Methods Network (AQMeN) in 2011 and 2013. The authors are also grateful to

the Scottish Government for their help in data matching.


3/22

I IntroductionThe financial crisis, and ensuing recession, has focussed policy makers attention onthe role of voluntary and civic participation in the provision of public services. This has

led to greater funding scrutiny and a need to understand how communities,individuals and volunteers can participate in, shape and support sustainable publicservices. This policy shift towards a co-production agenda assumes that levels ofvolunteering and its distribution in all localities can meet this demand now and in thefuture. However, voluntary participation varies across space, and across thedemographic and socio-economic characteristics of communities. In order tounderstand the impact that this shift in public service provision will have it is necessaryto understand the determinants of volunteering participation, and explain itsgeographical distribution.

In this paper data from five years of the Scottish Household Survey (SHS) arecombined with local area statistics from the Scottish Index of Multiple Deprivation(SIMD) and local aggregate measures of charitable activity from the Scottish CharityRegister in order to explore the determinants of volunteering in both the demand andsupply sides. This project explores variations of volunteering participation acrossspace, and in particular investigates the significantly higher levels of volunteeringparticipation observed in rural Scotland.

The question addressed here is: does volunteering participation vary significantly in

intensity and character between local areas, once the characteristics of individuals andcommunities have been taken into account?

The findings show that voluntary participation is significantly higher in more rural andmore remote areas, even after controlling for the demographic and socioeconomiccharacteristics of local areas. While these urban/rural differences in participationrates are significant, there is no significant difference in the number of hours ofvolunteering once local area differences are controlled for. Greater voluntary activityin rural areas is characterised by greater numbers of people being involved, ratherthan by more intensive participation of individual volunteers.

This persistence of spatial differences in volunteering participation presents bothchallenges and opportunities for policy makers. It suggests that there may beuntapped potential to encourage members of urban communities to increase theirinvolvement in voluntary activity. However, it also suggests that the burden of policymoves towards co-production and voluntary involvement in public service provisionmay fall disproportionally on some communities.

The rest of this paper is organised as follows. Firstly, there is a discussion of the

literature on the determinants of volunteering supply. Secondly, the data andmethods used are outlined, and the characteristics of volunteering in Scotland are

1


4/22

described in order to address the possible explanations for geographical differences.Thirdly, individual level data is used to control for demographics, a mixture ofindividual and local area data to control for socio-economic characteristics, and localarea data to partly control for volunteering demand. Lastly, the implications of these

findings are discussed, and the paper concludes.

II Literature on Voluntary ParticipationUnderstanding how the characteristics of individuals contribute to their propensity tovolunteer can help explain why participation might vary between communities. Whatdetermines the decision by an individual to volunteer? This question has beenaddressed most widely by sociologists, economists, and political scientists. Theexisting literatures focus on the effect that individual characteristics have on

volunteering propensity, but lacks a consistent theory of the volunteer. Thishighlights the complexity of the different motivations for volunteering, and the keythemes are discussed here.

When considering the social conditions that determine volunteering the mostcommon understanding is the human capital approach. This approach differs slightlywithin each discipline as each defines volunteering differently. Economics seesvolunteering as unpaid work and impure altruism; Sociology - social cohesion andsocial welfare; Political Science - citizenship and democracy. Furthermore, there arealternative paradigms through which volunteering can be viewed (Rochester, Paine,

Howlett, & Zimmeck, 2010). The dominant paradigm describes volunteering as analtruistic act to help others, usually performed in the context of an organisation orgroup. The civil society paradigm treats voluntary participation as being rooted in self-help and mutual aid, emphasising collaboration and working together. A thirdparadigm, volunteering as serious leisure, as an activity undertaken for enjoyment butwhich has a prosocial purpose.

It is widely argues that higher social and economic status is associated with volunteerparticipation (Wilson, 2000). This is supported by various arguments. Loseke (1997)argues that individuals with high financial and human capital status have a greatersurplus to offer as volunteers. High socioeconomic status individuals are more likely tojoin groups and organisations (including attending church) and so formal volunteeringis just one example of this behaviour (Farmer, 2006; Jones, 2006; Wilson & Janoski,1995). Rather than the broad conceptualisation of human capital, a large literaturefocuses on the determinants of human capital such as education, income, age, gender,marital status. There is also a literature that considers the direct role of socialnetworks which can be seen as a component of human capital. Each will be brieflyoutlined below.

McPherson & Rotolo (1996) has argued that education is the most consistentpredictor of volunteering. As years of education increases, the likelihood of

2


5/22

volunteering increases (Hustinx, Cnaan, & Handy, 2010) and many arguments havebeen put forward to explain this, from education increasing self-confidence (Brady,Verba, & Schlozman, 1995) and education increasing the broad measure of humancapital. It is widely accepted that income affects the supply of volunteering, often

based on the above argument that increasing financial capital permits morevolunteering (Menchik & Weisbrod, 1987). However this picture is not so clear, as ahigher income increases the opportunity cost of unpaid work; Freeman (1997) hasfound a negative relation between wage income and volunteering.

Marital status has been shown to impact on the decision to volunteer. A marriedindividual is more likely to volunteer (Wilson, 2000) particularly if their spouse alreadyvolunteers (Freeman, 1997). The presence of children is more complicated thanmarital status. Young children put a constraint on volunteering (Damico, Damico, &Conway, 1998) but overall the presence of children at home makes volunteering for

the parent more likely (Wuthnow, 1998).

Beyond the demographic and socio-economic characteristics of individuals, there arelarge variations in volunteering participation across space. Volunteering is often foundto be more common in rural environments (Chase-Ziolek & Striepe, 1999; Timbrell,2006). While this may be explained by differences in demographics between urbanand rural areas, there are also a number of theories as to why participation might varyby geography. Most discussions of the rural/urban distinction for volunteering focuson the different socioeconomic characteristics for each area. Rural communities showstronger and broader social capital (Hurtado, Kawachi, & Sudarsky, 2011; Ziersch,Baum, Darmawan, Kavanagh, & Bentley, 2009), increasing the potential forvolunteering in the rural communities.

Trimbell (2006) shows that the nature of volunteering differs between urban andrural environments. Rural volunteering tends to be broad, with more peoplecontributing across multiple organisations, but for few hours per week. Urbanvolunteering is deep, with fewer individuals working with fewer organisations but witha greater time commitment per week.

An alternative explanation for higher volunteering in rural areas comes from theeconomics literature, suggesting that volunteering participation will be higher in ruralareas due to the higher transaction costs of providing public services (Valentinov,2009). In more remote areas where the costs of providing public services is too highthere is an incentive for communities to cooperate locally through voluntaryorganisations in order to provide services.

The volunteering literature predicts that participation will vary across demography,socio-economics, and local area characteristics, but proposes a range of explanationsfor these variations. Differences in participation between communities may be

explained by differences in the characteristics of the members of those communities.Next, is an exploration of the extent to which the characteristics of local communities

3


6/22

have an impact on individual volunteering participation, beyond the characteristics ofthe individual and household. In doing this, a better understanding of regional andlocal variation in participation is developed, and the consequences that this mighthave for policies promoting greater reliance on voluntary action.

III Data and MethodologyThe survey data used in this paper is from the Scottish Household Survey (SHS)between 2007 and 2011. The SHS is a continuous survey of households in Scotland,sampling around 31,000 households over a two year period. The module includingquestions on volunteering is asked of half the SHS sample.

Question design

As not everyone shares the same definition of what constitutes volunteering thewording of a volunteering question is very important. The volunteering questions inthe SHS have undergone periodic revision, making comparisons across longer timeperiods difficult. The questions have been stable since 2007, with only minoramendments to categories. The questions are asked of one random adult aged over16 within the household.

The SHS captures volunteering in the past 12 months through a two-stage question.This two-stage approach tackles some of the problems in identifying activities that

respondents may not think of as volunteering. In asking about unpaid help for groupsand organisations this question captures participation in formal volunteering. It wouldhowever not necessarily capture involvement in informal volunteering that takes placeoutside the context of a group or organisation.

Respondents who answer yes to either of these questions are coded as havingvolunteered in the past year. They are then asked a succession of more detailedquestions about the types of volunteering they have undertaken.

Volunteers are also asked about the number of hours they have spent in the previous

month volunteering. This question provides a measure of intensity of voluntaryparticipation.

Administrative data was then matched to the individual survey data in order toprovides measures of the characteristics of local areas. This administrative data isdrawn from the Scottish Index of Multiple Deprivation and from the Scottish CharityRegister.

Local area data from the Scottish Index of Multiple Deprivation (SIMD) was matchedto individual SHS records by statistician in the Scottish Government. The ScottishIndex of Multiple Deprivation combines 38 indicators across 7 domains, namely:

4


7/22

income, employment, health, education, skills and training, housing, geographic accessand crime. The income domain combines measures of the number of adults andchildren in households receiving income support, pension credits, job seekersallowance and tax credits. The employment domain includes the unemployment

claimant count, recipients of incapacity and disablement benefits. The health domaincombines a range of health indicators including mortality ratio, alcohol and drug-related hospital episodes, mental health, and low-birth-weight births. The educationdomain covers school pupil performance and absence, higher education enrolment,and numbers with no qualifications. The housing domain includes a measure ofovercrowding, and households without central heating. The geographic access toservices domain combines drive times and public transport times to GPs, schools, andshopping and post offices. The crime domain includes crime rates for house breaking,drug offences, assault and vandalism.

Five of the domains (income, employment, health, housing, geographic access toservices) are included in the model as measures of need, and used as proxies for thedemand for volunteers. Two of the domains (education, crime) are included asproxies for social capital in communities. The inclusion of these fine measures ofdeprivation across seven domains enables the inclusion of a basket of proxies forvolunteering demand in local areas.

The Scottish Charity Register is a register of all charities in Scotland maintained by theOffice of the Scottish Charity Regulator (OSCR). Any organisation wishing to operateas a charity must apply to OSCR for inclusion in the Register. The Register containsdetails of 20,000 Scottish charitable organisations, including their location, financialdata, beneficiary groups and charitable purposes.

Data Merging

For this project assistance was received from the Scottish Government to merge inthe local area data, matched to individual survey records at the data zone level. Datazones are groups of Census output areas, covering the whole of Scotland and nestedwithin local authority boundaries, which have populations of between 500 and 1,000

household residents. In addition, they have compact shape and contain householdswith similar social characteristics. There are 6,505 data zones across Scotland, they area stable geography and can be used to analyse change over time.

The SIMD statistics within each domain are combined to calculate a score, anddatazones are then ranked by these scores. The variables for each domain in thispaper are centiles of the datazone ranking.

The Scottish Charity Register was used to produce aggregate statistics of the numberof charities for each beneficiary group and charitable purpose and these were

5


8/22

produced at the intermediate zone level. Intermediate zones are groups of datazones covering small areas, each containing at least 2,500 individuals.

This set of local area statistics at the lower data zone level and higher intermediatezone level was matched by Scottish Government analysts to individual respondents inthe survey data based on their home address. No identifying information wasprovided to the author.

Descriptives

Descriptive statistics on volunteering in the SHS have been produced by the ScottishGovernment (2007; 2009). Some basic descriptives are first described here, beforeproceeding to a more in-depth analysis of the factors affecting volunteeringparticipation.

Around a third of adults in Scotland say that they have volunteered in the past 12months, as shown inTable 1. This figure has held fairly steady for several years.Participation is slightly higher for women (31.8%) than men (28%). Volunteeringparticipation increases significantly as rurality increases, and also suggests that itincreases as accessibility decreases from as low as 25.7% for men in large urban areasto as high as 51.8% for women in very remote rural areas.

Table 1: Volunteering Participation by Rurality &

GenderOn average male volunteer spend about 11.1 hours and female volunteers 9.7 hoursvolunteering in a month. Hours spent volunteering are highest for both men andwomen in very remote rural areas, but the difference is not very significant.

Methodology

The analysis begins by firstly examining the factors influencing the participation

decisions. Secondly, it examines the factors affecting intensity of participation,measured by the number of hours volunteered.

Volunteer participation is estimated by a binary logistic regression of volunteering ona range of individual, household and geographical characteristics. The regression ofvolunteering hours is estimated by Ordinary Least Squares. The key test in theseregressions is the extent to which an urban/rural difference in participation remainsonce individual and community characteristics are controlled for.

IV Results

6


9/22

Volunteering Participation

Table 2 shows the estimation results for the logistic model of volunteeringparticipation. Women are more likely to volunteer than men. There is very little

variation in volunteering probability across the age categories, with participationlowest for those aged 25 to 39 years, and not significantly declining in later life untilover 75 years. Education is a significant predictor of volunteering, with participationincreasing with the level of individuals highest qualifications. Income is positivelycorrelated with volunteering participation, through both home ownership andhousehold income. Individuals in couples are less likely to volunteer than those livingalone. Having children makes individuals more likely to volunteer than those without.Reported religious affiliation is positively correlated with volunteering participation.

Individuals in rural areas are significantly more likely to volunteer than those living inmore urban areas. Within the urban/rural category, participation also increases withremoteness, with individuals in the most remote rural areas 2.3 times more likely tovolunteer than those in large urban areas. Of the seven SIMD dimensions three arestatistically significant. Individuals living in areas with higher levels of education, andthose living in areas with better health, are more likely to volunteer. Perhapssurprisingly, we find that individuals living in lower income areas are more likely tovolunteer, given their other characteristics. As the positive effect of householdincome on volunteering participation is controlled for, there is a relative incomeeffect; e.g. for a given level of household income and individual living in a more

deprived area is more likely to volunteer than an individual living in a wealthier area.This suggests that relative income is important; it is the relatively wealthy that aremost likely to volunteer.

Controlling for the local area characteristics does have a small downward effect on therurality coefficients, suggesting that local area differences explain part of the ruralityeffects. Controlling for the characteristics of locally-based charitable organisationexplains even more of the urban/rural difference in participation, with individuals inthe most rural areas now only 1.7 times more likely to volunteer than those in urbanareas. However, significant effects of rurality on participation remain.

Table 2: Logit Regression of Volunteering Participation

Volunteering Hours

Table 3 contains the estimation results for volunteering hours. There is no significantgender difference in number of hours volunteered. Having children significantlyreduces the number of hours spent volunteering by around 2 hours per month, withgreater numbers of children having a bigger effect. Age has little effect on

volunteering hours until post-retirement age; those aged 65 years and over volunteersignificantly more hours, peaking with those aged 60-69 years giving an additional 4.5

7


10/22

hours per month on average. There is no significant link between income and hoursspent volunteering, although individuals outside the labour market are likely tovolunteer longer hours. Reported religion is positively associated with volunteeringhours, as is education.

No evidence of any effect of rurality or accessibility on the number of hoursvolunteered is found. Similarly, there is little impact found on volunteering hours ofcontrolling for local area characteristics.

Table 3: OLS Regression of Volunteering Hours

V DiscussionAs described in the volunteering literature, the analysis shows that individual

characteristics such as gender, age, and education are important in determiningvolunteering participation. Household characteristics such as relationship status andnumber of children also have a significant effect. The characteristics of theneighbourhoods in which people live then have a further impact on participation. Thisanalysis has shown that even after controlling for a wide range of individual andhousehold characteristics there still remains a sizable urban/rural difference inparticipation. Furthermore, controlling for local differences in neighbourhoodcharacteristics reduces but does not eliminate this urban/rural effect.

The social benefits of volunteering are suggested by the higher participation of thoseliving alone. From an opportunity cost perspective the higher participation ofindividuals with children is surprising; however it seems likely that having childrenprovides more opportunities to volunteer, and that this social network effectdominates. The opportunity cost of volunteering when individuals have children canbe seen in the negative effect that having children has on the number of hourscommitted to volunteering.

Higher income and education groups have higher volunteering participation rates. Inaddition, the levels of health and education in the local area also have positive impact

on participation. Within the model this could be explained by both lower search costsof finding volunteer opportunities through higher levels of human and social capital.However, it should be noted that some research (F. M. Smith, Timbrell, Woolvin,Muirhead, & Fyfe, 2010) suggests that there are greater levels of informalvolunteering in more deprived areas and that some of these activities may not bepicked up in the survey.

Volunteering participation is significantly higher in rural Scotland than urban areas.This is the case even after controlling for the income and demographic differencesthat may be present, and is despite potentially greater costs in travel. Those in themost remote rural areas have even higher participation still.

8


11/22

The difference in urban/rural volunteering participation is slightly reduced but noteliminated by controlling for a range of local area characteristics. This suggests thatthe participation differences are not explained either by differences in demographicsbetween urban and rural communities, or in differences in community characteristics

such as health, education or deprivation. Controlling for the numbers and types ofcharities based locally has a much greater effect in explaining the urban/ruraldifferences in participation, but does not remove the difference completely.

There does not however seem to be an urban/rural difference in the number ofhours spent volunteering. This suggests that the significant difference between urbanand rural areas lies in the decision to participate, rather than in the intenstiy ofparticipation. So while a much greater proportion of the population in rural areas arevolunteering, individually they are not volunteering any more hours.

VI ConclusionThis paper has explored the extent to which a range of demographic and economicvariables can explain volunteering; both the decision to participate and the extent ofparticipation. It has shown that volunteering in Scotland is in line with many of thetheoretical predictions and empirical findings from other volunteering studies. Ofparticular interest in the Scottish context is the urban/rural difference in participation,which remains even after many other observable characteristics have been controlledfor. The contribution of this paper has been to attempt to control for local area

characteristics in order to explain the geographical variation in participation. Evenafter controlling for these local measures of need, significant differences inurban/rural participation remain. Through examining the extent of participation ithas shown that this urban/rural difference is present only in the decision toparticipate, and does not appear to affect the scope of involvement.

The findings suggest that there is a dual explanation for higher rural volunteeringparticipation: greater levels of social capital; and increased supply of voluntaryopportunities linked to service provision. Both of these factors play a role in thedecisions of individuals to volunteer.

The geography of volunteering suggests that the supply of volunteering varies acrossspace, dependent on both the characteristics of the local population and the localarea. As such, policies to increase voluntary participation in public service provisionare likely to be more successful in some areas than others. This could occur in one oftwo ways. Communities with existing higher levels of participations may be betterequipped to absorb additional voluntary roles, as the social connections and supportstructures are already in place. Alternatively, these communities may already be atcapacity, with available volunteers unable to take on any more responsibility for public

service provision. More urban communities may have untapped potential but weakernetworks, while more rural communities have better connections but risk overloading

9


12/22

individuals. Regardless, different approaches will need to be taken by policy makersthat take account of the local context if volunteering is to be supported; and if theimpacts of cuts in formal service provision are not to fall disproportionally on somecommunities.

Identifying knowledge gaps which if addressedcould better support policy and practicedevelopment1. Exploring spatial variation of volunteering participation in othergeographies

While there is a growing body of evidence in Scotland exploring the differences in the

nature and extent of volunteering between urban and rural communities, thoseworking locally local government, community planning partners, Third SectorInterfaces and so on - need a greater understanding of volunteering participationwithin and between other geographies like local authority areas in order to makeinformed decisions about volunteer development.

2. Improving how volunteering is measured in Scotland

Understanding volunteering participation in Scotland through statistical means iscurrently limited as the household survey captures only formal volunteering throughorganisations or groups. An informal measure of volunteering would allow for a morein depth and nuanced understanding of the nature and extent of volunteeringparticipation. In addition, it would more adequately recognise and value the diversityof voluntary activity across all individuals and communities in Scotland.

10


13/22

ReferencesBrady, H. E., Verba, S., & Schlozman, K. L. (1995). Beyond ses: A resource model of politicalparticipation.The American Political Science Review, 89(2), pp. 271-294.

Chase-Ziolek, M., & Striepe, J. (1999). A comparison of urban versus rural experiences of

nurses volunteering to promote health in churches.Public Health Nursing, 16(4), 270-279.

doi:10.1046/j.1525-1446.1999.00270.x

Damico, A. J., Damico, S. B., & Conway, M. M. (1998). The democratic education of women:

High school and beyond.Women & Politics, 19(2), 1-31.

Farmer, G. (2006). African american males civic engagement: The importance of social

organization involvement and friendship diversity.Journal of African American Studies, 10(2),

51-68. doi:10.1007/s12111-006-1003-6

Flanagan, C. A., Bowes, J. M., Jonsson, B., Csapo, B., & Sheblanova, E. (1998). Ties that bind.

Journal of Social Issues, 54(3), 457-475. doi:10.1111/j.1540-4560.1998.tb01230.x

Freeman, R. B. (1997). Working for nothing: The supply of volunteer labor.Journal of Labor

Economics, 15(1, Part 2: Essays in Honor of Yoram Ben-Porath), pp. S140-S166.

Gaskin, K., & Volunteer Centre UK. (1995). In Smith J. D. (Ed.), A new civic europe? : A study of

the extent and role of volunteeringVolunteer Centre UK.

Hurtado, D., Kawachi, I., & Sudarsky, J. (2011). Social capital and self-rated health in colombia:

The good, the bad and the ugly.Social Science & Medicine, 72(4), 584-590.

doi:10.1016/j.socscimed.2010.11.023

xi


14/22

Hustinx, L., Cnaan, R. A., & Handy, F. (2010). Navigating theories of volunteering: A hybrid

map for a complex phenomenon.Journal for the Theory of Social Behaviour, 40(4), 410-434.

doi:10.1111/j.1468-5914.2010.00439.x

Jackson, E. F., Bachmeier, M. D., Wood, J. R., & Craft, E. A. (Spring 1995). Volunteering

and charitable giving: Do religious and associational ties promote helping behavior?

Nonprofit and Voluntary Sector Quarterly, 24(1), 59-78. doi:10.1177/089976409502400108

Jones, K. S. (2006). Giving and volunteering as distinct forms of civic engagement: The role

of community integration and personal resources in formal helping.Nonprofit and Voluntary

Sector Quarterly, 35(2), 249-266. doi:10.1177/0899764006287464

Karniol, R., Grosz, E., & Schorr, I. (2003). Caring, gender role orientation, and volunteering.

Sex Roles, 49(1), 11-19. doi:10.1023/A:1023953401662

Loseke, D. R. (1997). The whole spirit of modern philanthropy: The construction of the idea

of charity, 1912-1992.Social Problems, 44(4), 425-444.

Marwell, G. (1993). In Oliver P. (Ed.), The critical mass in collective action : A micro-social

theory. Cambridge: Cambridge University Press.

McPherson, J. M., Popielarz, P. A., & Drobnic, S. (1992). Social networks and organizational

dynamics.American Sociological Review, 57(2), pp. 153-170.

McPherson, J. M., & Rotolo, T. (1996). Testing a dynamic model of social composition:

Diversity and change in voluntary groups.American Sociological Review, 61(2), pp. 179-202.

Menchik, P. L., & Weisbrod, B. A. (1987). Volunteer labor supply.Journal of Public

Economics, 32(2), 159-183. doi:DOI: 10.1016/0047-2727(87)90010-7

xii


15/22

Midlarsky, E. (1994). In Kahana E. (Ed.),Altruism in later life. Thousand Oaks ; London:

Sage.

Rochester, C., Paine, A. E., Howlett, S., & Zimmeck, M. (2010). Volunteering and society in

the 21st centuryPalgrave Macmillan Hampshire.

Schoenberg, S. P. (1980). Some trends in the community participation of women in their

neighborhoods.Signs, 5(3, Supplement. Women and the American City), pp. S261-S268.

Smith, D. H. (Fall 1994). Determinants of voluntary association participation and

volunteering: A literature review.Nonprofit and Voluntary Sector Quarterly, 23(3), 243-263.

doi:10.1177/089976409402300305

Smith, F. M., Timbrell, H., Woolvin, M., Muirhead, S., & Fyfe, N. (2010). Enlivened

geographies of volunteering: Situated, embodied and emotional practices of voluntary action.

Scottish Geographical Journal, 126(4), 258 274.

doi:10.1080/14702541.2010.549342

Taniguchi, H. (2006). Men's and women's volunteering: Gender differences in the effects of

employment and family characteristics.Nonprofit and Voluntary Sector Quarterly, 35(1), 83-

101. doi:10.1177/0899764005282481

Timbrell, H. (2006). Scotlands volunteering landscape: The nature of volunteering.

Unpublished manuscript.

Valentinov, V. (2009). Third sector organizations in rural development: A transaction cost

perspective.Agricultural and Food Science, 18(1), 3-15. doi:10.2137/145960609788066825

Wilson, J. (2000). Volunteering.Annual Review of Sociology, 26(1), 215.

xiii


16/22

Wilson, J., & Janoski, T. (1995). The contribution of religion to volunteer work.Sociology of

Religion, 56(2), 137-152. doi:10.2307/3711760

Wilson, J., & Musick, M. (1997). Who cares? toward an integrated theory of volunteer work.

American Sociological Review, 62(5), pp. 694-713.

Wuthnow, R. (1998).Loose connections: Joining together in america's fragmented

communities. Cambridge: Harvard University Press.

Ziersch, A. M., Baum, F., Darmawan, I. G. N., Kavanagh, A. M., & Bentley, R. J. (2009).

Social capital and health in rural and urban communities in south australia. Australian and

New Zealand Journal of Public Health, 33(1), 7-16. doi:10.1111/j.1753-6405.2009.00332.x

xiv


17/22


18/22

Table 2: Logit regression of volunteering participationModel 1:

Individual DataModel 2:

Matched local area dataModel 3:

Charity data

female

1.135 1.134 1.135

(0.0454)*** (0.0456)*** (0.0456)***

Children: 1-2

1.255 1.241 1.250

(0.0687)*** (0.0683)*** (0.0689)***

Children: 3-4

1.266 1.246 1.250

(0.137)** (0.135)** (0.136)**

Area: Other urban1.000 1.004 0.997

(0.0463) (0.0529) (0.0534)

Area: Small accessible towns

1.124 1.110 1.081

(0.0761)* (0.0795) (0.0788)

Area: Small remote towns

1.194 1.137 1.083

(0.127)* (0.124) (0.121)

Area: Very remote small towns

1.635 1.674 1.518

(0.219)*** (0.227)*** (0.221)***

Area: Accessible rural

1.298 1.224 1.168

(0.0769)*** (0.0852)*** (0.0865)**


19/22

Area: Remote rural

1.961 1.810 1.648

(0.186)*** (0.185)*** (0.182)***

Area: Very remote rural

2.261 2.079 1.682

(0.194)*** (0.207)*** (0.220)***

SIMD: Crime

0.971 1.021

(0.0964) (0.103)

SIMD: Geographical Access to Services

0.979 0.942

(0.0856) (0.0830)

SIMD: Housing

0.994 1.062

(0.108) (0.119)

SIMD: Education & Skills

2.055 1.934

(0.313)*** (0.300)***

SIMD: Health

1.622 1.470

(0.292)*** (0.265)**

SIMD: Employment

1.120 1.051

(0.249) (0.235)

SIMD: Income

0.443 0.457

(0.104)*** (0.108)***

Charity Beneficiary Groups -


20/22

Charitable Purposes -

N

22970 22970 22970

AIC

25218.2 25146.8 25122.2

Exponentiated coefficients; Standard errors in parentheses

*p< 0.10, **p< 0.05, ***p< 0.01

Other variables included in the regressions but not reported above include house ownership, marital status, number of children, disability,household work status, household income, occupation, education, religion, ethnicity.

Observations are weighted using the Scottish Household Survey sample weights.

(Source: Scottish Household Survey 2007 to 2011)


21/22

Table 3: OLS Regression of Volunteering HoursModel 1:

Individual Data

Model 2:

Matched local area data

female -0.743 -0.706(0.598) (0.599)

Children: 1-2 -1.443 -1.363

(0.816)* (0.816)*Children: 3-4 -2.072 -1.856

(1.198)* (1.198)Area: Other urban -1.046 -0.505

(0.664) (0.717)

Area: Small accessible towns 1.098 1.580(1.513) (1.569)

Area: Small remote towns 0.580 0.901

(1.549) (1.571)Area: Very remote small towns -1.320 -1.069

(1.242) (1.344)Area: Accessible rural -1.234 -0.200

(0.748)* (0.927)Area: Remote rural -0.524 0.686

(1.163) (1.315)

Area: Very remote rural 1.263 2.396(1.127) (1.339)*

SIMD: Crime - -1.799(1.464)

SIMD: Geographical Access to Services - 1.097

(1.423)

SIMD: Housing - -0.899(1.462)

SIMD: Education & Skills - 3.375(2.349)

SIMD: Health - -0.702(2.711)

SIMD: Employment - -1.224(3.681)

SIMD: Income - -0.356(3.511)

_cons 8.658 8.146(2.932)*** (3.360)**

N 5361 5361R2 0.047 0.049

AIC 45095.9 45099.4

Standard errors in parentheses

*p< 0.10, **p< 0.05, ***p< 0.01

Other variables included in the regressions but not reported above include house ownership, marital status, number of children, disability,household work status, household income, occupation, education, religion, ethnicity.

Observations are weighted using the Scottish Household Survey sample weights.

(Source: Scottish Household Survey 2007 to 2011)


22/22

Session 3B Doyle, Rutherford, McDonnell

Documents

Transcript of Session 3B Doyle, Rutherford, McDonnell